Uncertainty in the Variational Information Bottleneck

Abstract

We present a simple case study, demonstrating that Variational Information Bottleneck (VIB) can improve a network's classification calibration as well as its ability to detect out-of-distribution data. Without explicitly being designed to do so, VIB gives two natural metrics for handling and quantifying uncertainty.

10 pages, 7 figures. Accepted to UAI 2018 - Uncertainty in Deep Learning Workshop